Serious question here, Why should I buy a Ryzen CPU if the 7700K

Why should I buy a 7700k when a 1700 is cheaper and I can just turn off cores and OC when I need higher freqs and use full 8 cores when I'm compiling gentoo?

>>59149010
I don't know but you're an idiot

>>59148945
If you think 4c/8t is more than enough, then you're probably going to be interested in the Ryzen R3 series of CPUs. They'll be at least half the price of the i7-7700K with similar IPC. The i7-7700K will overclock a bit higher but cost a lot more. For most, I think the R5 series (6c/12t) will be the sweet spot.

Also bear in mind that in a lot of games, Intel's lower-clocked and previous-generation 6c/12t CPUs out-perform the i7-7700K so the era of quad cores being optimum for gaming is pretty much at an end.

File: smug_cartoon_face.jpg (27KB, 279x304px) Image search: [Google]

27KB, 279x304px

>>59149042

Sorry, but you guys here and the rest of you on the internet will get a rough wake up on the 2nd March.

OC is feasible if you have a high quality water cooling system, and then the reviewer told be me that going beyond the Turbo clock + XFR will not really pay out.

The Ryzen 1700/1800 more than anything else is a multitasking CPU and as such intended for people who need it.

>>59149010
>hey /g/ why should I buy X if Y is cheaper
>because X is cheaper

this is you

>>59149100
>the reviewer
Who?

I'm going for Ryzen because I'm going to stream my gaming, so the extra cores come in handy.

just bought a 7700k

gayzen's only competitive chips are still priced way over kaby lake anyway.

>>59148945
Because the 1700x is 8 core and alot of programs are decently multi threaded now

(even gaymes)

>>59149148
>just bought

Idiot, I also will most probably buy an i7, but only after Intel cuts the prices.

File: [Continued laughter].gif (4MB, 273x207px) Image search: [Google]

4MB, 273x207px

>>59149148
>T-t-tell me it'll b-be okay, /g/...!

File: 1484857165b9Q0qmS6fl_1_6.png (13KB, 620x336px)

13KB, 620x336px

Have fun destroying your cpu while trying to delid it because intel couldn't be arsed to solder their IHS to their flagship i7.

>>59149177
8 cores and 16 threads is still overkill for the absolute majority of people.

Don't make me laugh by trying to justify that the Ryzen 7 series is actually pandering towards the mainstream market.

>>59149228
well the majority of people are going to have 4 core apus

It fucking bothers me that we won't get any reviews of the 1600X and the 1600 any time soon.

To hell with the NDA.
If I knew that the 1600X would be almost i7-7700 performance, runs cooler and drains less power, I would even wait for another 3 months.

>>59149219
They do it intentionally to save a tiny amount of money, give their HEDT platform more selling points, and to allow resellers to do "delidded bundles" at inflated prices.

>>59149294
The only thing we won't know about the R3 and R5 CPUs by 2nd March is how high they overclock.

>>59149294
You will know tomorrow. The 1600x has the same cores and clock speeds as the 1800x, so it will be about the same performance on single threaded applications.

>>59149323
Even that CAN be found out by disabling 2 cores in Ryzen Master. Wonder how many will do this kind of extensive testing though.

>>59149228
just buy a 4 core ryzen and shut the fuck up

>>59148945
Power consumption?

I dunno but it's an exciting chip, shame Lightroom and Photoshop won't benefit from all those corns.

Just bought an i7 7700K.

Fuck AMD, and fuck Poozen.

>>59148945
>4C/8T is more than enough
If you believe that and you want to pay out the ass for your CPU, buy it. Otherwise hop on the Ryzen express now or wait for Series 5/3

>>59149360
>i like to be cucked
are you swedish?

Just bought an i7 7700K.

Screw AMD, and fuck Poozen.

>>59149055
R5 has 6c and 4c cpus as well.

Because the 1700 competes with Broadwell-E and the upcoming Skylake-E, not Kaby Lake. The Ryzens you SHOULD be considering are the hexa and quadcores that come out later in the spring.

File: laugh.jpg (8KB, 209x200px)

8KB, 209x200px

>>59149382
>You couldn't even wait for price drops
>Brand Loyalty
It's like you want to get fucked in the ass.

>>59148945
Probably because nothing uses single threaded processes anymore.

Just Purchased an i7 7700K.

Fuck AMD, and fuck Poozen.

>>59149382
>>59149125
>The only reason someone would want AMD to fail is if they work for Intel, or if they just bought an i7 and are having buyers remorse, but even then they should be happy that there will be a leap in progress and future upgrades will be cheaper.
Guess you're the latter.

>>59149356
power consumption is and will always be better with Intel, I doubt a 65W Ryzen CPU will stay within the 65W range under a gaming load. They still haven't updated the way they measure TDP, so a 65W Ryzen CPU will reach 100W easily.

>>59149055
it should be noted that the margin these handfull of game are "outperformed" by is rarely more than 5 fps.

>>59149442
True but it shows Intel's MOAR MHZ doesn't always win any more.

>>59149219
as long as you're not using a stock cooler there's literally no danger in overclocking kaby, you don't have to delid it and this meme is just a fucking meme.

>>59149433
>TDP is power consumption

>>59149148
buyers remorse purchase

OP here

so, all this time /g/ has been telling me to wait for Ryzen because it'll be cheaper, but now that it's released and it's actually more expensive than Intel's alternatives, I should buy it because it has more cores? Huh...so that's still AMD's move, "we have more cores"?

>>59149466
>you don't have to delid it and this meme is just a fucking meme.
You do if you want good temperatures.

>>59149458
it does if you test overclock vs overclock in 4 core configurations.

>>59149481
you're an idiot

File: laughing harpies.gif (729KB, 416x234px) Image search: [Google]

729KB, 416x234px

>>59149433
>he thinks TDP refers to power consumption

File: 123123.jpg (336KB, 1131x817px)

336KB, 1131x817px

eyzen 1700x faster than 7700k in single thread and faster than 6950k in multithread
sauce:
http://valid.x86.fr/bench/rjmzdu/16
http://valid.x86.fr/bench/rjmzdu/1

https://www.reddit.com/r/Amd/comments/5wg020/1700x_single_multithreaded_cpuz_benchmark_valid/


REKT

>>59149487

Just spent my money on an i7 7700K.

Fuck AMD, and fuck Poozen.

>>59149489
there's no difference in the longevity of your hardware at 80 degrees vs 60 degrees.

>>59149148
>>59149408
>>59149382
>denial/bait

>>59149487
t. intelshill

Only the highest end ryzens have been released we still have to wait for low end ones :^)

>>59149487
But it will be cheaper, you fucking retard. The 1700 isn't cheaper than 7700k because it doesn't compete with 7700k, it competes with 6900k and 6850k.

File: tb.jpg (6KB, 183x276px)

6KB, 183x276px

4 cores in 2017

>>59149466
>literally see temps drop 13 fucking Celsius when stock jizzed TIM is replaced
>meme
>meme
>meme
Holy shit, how much is intel paying you?

>>59149512
test it stock now

File: tg551xu3z1iy.png (97KB, 768x830px) Image search: [Google]

97KB, 768x830px

>>59149527
>>>59149487
>But it will be cheaper, you fucking retard. The 1700 isn't cheaper than 7700k because it doesn't compete with 7700k, it competes with 6900k and 6850k.
SIR PLS DELET

File: undyne smug.png (215KB, 500x466px) Image search: [Google]

215KB, 500x466px

>>59149512
>tfw waited for ryzen

feels good man

>>59149529
In 2017 you literally need both a quadcore and an 8 core

Although a 6 core is a good compromise between them

>>59149532
with no information based on cooler or load. if the load is artificial the test wont matter much in the real world anyway. it is a fucking meme, you don't need to change your tim, you don't have to delid your processor. there is no difference between running your cpu at 80 degrees constantly and 60 degrees and under constantly.

>>59149010
>I can just turn off cores and OC when I need higher freqs
No you can't. That's not how it works you fucking retard

>>59149581
>there is no difference between running your cpu at 80 degrees constantly and 60 degrees and under constantly.
I really hope you were just trolling me this whole time anon-kun...

>>59149590
except it does in ryzen master

File: AMD-Ryzen-Master-Overclocking-Utility[1].png (2MB, 2560x1325px) Image search: [Google]

2MB, 2560x1325px

>>59149590
on a scale of "shill" to "shill btfo", where are you on the scale now?

>>59149581
>there is no difference between running your cpu at 80 degrees constantly and 60 degrees and under constantly.

>silicon will not degrade faster

ENJOY YOUR CPU ONLY LASTING 10 YEARS FGT LOL

>>59149544
U are dying inside because intel cucked u and u can't take it, stop fighting on internet stop defending evil.

https://www.guru3d.com/news-story/intel-is-trying-to-manipulate-amd-ryzen-launch.html

>people are starting to compare a 440$ CPU to a 100$ less expensive CPU

Ok

>>59149512
overclocked 1700x vs stock 7700k when most chips are hitting 5.0ghz, which ryzen will not even come close to. also the 1700x is more expensive, and with intel price drops, the 1700x becomes even less appealing

>>59149590
Not with intel.

You can with Ryzen, so stop buying goyware.

>>59149613

but amd is the jewish company.

>>59149529
You literally don't need anything more

>>59149672
This, the human eye can't see past 4 cores.

>>59149601
>>59149604
>>59149634
You can turn off cores all you want but you won't be getting better OC speeds.

>>59149512
Most likely OC with LN2 but nice to see these numbers regardless

>>59149597
there is no difference. anyone at amd or intel will tell you, as long as your cpu is run within temperature specs it does not affect it's longevity.

the most risk of damage to your cpu in these scenarios would be overvolting while attempting to overclock.

>>59149611

overclocking will degrade your chip more than heat will. but thanks for not contributing anything remotely useful.

>>59149692
>it's nice to see that an OC'd $400 cpu can barely beat $320 @stock speeds
AMDrones are hilarious

>>59149690
oh, I thought you were talking about the ability to disable cores

sorry, I thought you were saying that it's not possible to disable individual cores at all

Just Purchased an i7 7700K.

Fuck AMD, and fuck Poozen.

>>59149690
Then why are literally all OC records done with all except one of the cores disabled?

>>59149382

nigga you could've wait for the inevitable Intel panic price drop. I'd get a 7700k too but 1151 mITX are overpriced like fuck, so I'll just wait for AM4 ones.

>>59149752
To make sure the VRMs can take it without blowing up.
But hey if you want to OC your 1700x to 5.0 GHz for a few minutes with LN2 I'm sure you can.
You won't be getting a stable OC on it past the boost speeds however

>>59149529
There's literally no point to having anything with more than 4C/8T in terms of gaming.

>>59149795
8t is already overkill actually

>>59149687
+1

>>59148945
I'm still getting a 1700x or 1800x. I got dozens of JAV that need to be converted from wmv to mp4/mkv.

>no Ryzen equivalent of the i7-7700K
>R3 and R5 have less cores
>mediocre IPC at best
In other words, completely useless for high-end gaming.

>>59149853
>needing the equivalent of jew shit

>>59149853
>mediocre IPC at best
except the leaks show that Zen cores have comparable or even better IPC to Kaby Lake, they just don't have the MHz

>muh gaming
Unless you have a meme 144hz monitor the 1700 is overkill for any game, so is the 7700k to be honest
I have an i5 4670 and I've yet to hit any cpu bottlenecks.

>>59149861
>both companies run by poojeets and isrealis
>one is seemingly more jewish than the other.

>>59148945
You do realize 8 real cores is better than 4c/8t even if the clocks are a bit higher right?

File: ryzen.png (1MB, 700x960px) Image search: [Google]

1MB, 700x960px

https://www.youtube.com/watch?v=9AVZ_x64hg4

>>59149921
not for gaming, or anything else other than video encoding

>>59149788
Less cores means less heat, can take more voltage and more voltage means more overclocking potential.
Ryzen engineering sample on a single core was overclocked to 5.1 on air.

>>59149921
better for certain usages*

>>59149935
Less cores only mean less voltage for the VRM to handle which means you can have a pretty unstable higher clock for a few minutes on LN2, enough time to run benchmarks and that's it

>>59149928
Well the 1700x and 1800x certainly aren't going to bottleneck current games, and future games will support more cores.

>>59149956
I didn't say less cores need less voltage, I less cores can take more voltage for more overcloking potential without getting toasted.

>>59149959
>the 1700x and 1800x certainly aren't going to bottleneck current games
They are unless you have a shitty 60Hz monitor, but if you do you better off waiting for the R3 or getting a 6600k now

>>59149921
>muh cores
Literally the only thing Ryzen has going for it. Unfortunately actual PC gamers don't give a shit about streaming/editing/whatnot and just want a CPU that won't bottleneck their GTX 1080.

>>59149974
And I'm saying it's all for the VRMs sake, not the cpu.
That's why they disable cores.

File: Ashes.png (13KB, 1097x697px)

13KB, 1097x697px

>>59149928
If a game uses 8 threads or more the Ryzen chip should be faster assuming it's fully utilizing 8 threads

like in ashes here, hopefully this becomes the norm for DX12/Vulkan titles

>>59149996
>Ashes of Benchmark
Please link Watchdogs 2 and BF1 after, should be somewhere on the graphs folder the marketing team gave you this week.

>>59149990
Which it won't? Unless you're trying to do the most extreme 144fps gaming in AAA games

File: 1421747998694.jpg (89KB, 800x1000px) Image search: [Google]

89KB, 800x1000px

>>59149996
>8 real cores are better than 8 fake ones

what a fucking shocker

time to dust off the old memes

>>59149996
one of 5 games that scales multicore and one of 16 dx12 games. that is a very limited use case to buy a cpu specifically for anon

>>59149795
Depends if you wanna have tasks running in the background.

>>59149996
lol DX12 has fuck all to do with CPUs, it's a graphics API that only benefits AMD GPUs

>>59149994
Cpu core heat also plays a role.

File: WD 2 8 threads.png (71KB, 806x696px) Image search: [Google]

71KB, 806x696px

>>59150017
BF1 only uses 4 threads, it can use 8 but it barely scales past 4

Watch Dogs 2 will be very interesting since it utilizes 8 threads well, the R7 should be faster than the 7700K here

>>59150034
Except this should be more common, and it's not like the Ryzen chip will get below 60fps in 99% of games since it it's finally on par with IPC

>>59150038
Your idiot

File: xBgSksM.jpg (18KB, 340x260px) Image search: [Google]

18KB, 340x260px

>>59150038

Not only is the main benefit of DX12 almost entirely cpu side everyone benefits because of that.

>>59150046
>Cpu core heat also plays a role.
No it fucking doesn't because they use LN2 so all the cores are at constant -100 temps

>muh games

>>59148945
You shouldn't.

The R7 Ryzen CPUs are garbage.

Literally the only reason to buy them is if you are doing lots of video encoding.

>>59150050
>it's not like the Ryzen chip will get below 60fps
>60hz
>buying a $400 cpu to get 60fps
kill urself

>>59150061
That post isn't talking about ln2 >>59149010. If you are on air disabling cores will will increase your OC potential on air

>>59150021
>Unless you're trying to do the most extreme 144fps gaming in AAA games
What's the point of building a gaming rig if you're not going to make use of the best hardware available? Otherwise just buy a cheap console for half the cost of a "budget" meme PC.

>>59150058
most dx12 supported games are still played on dx11 legacy because of the software concerns ms10 has. so you would have to sell your soul to reap any benefit.

>>59150038
>it's a graphics API that only benefits AMD GPUs

Fucking chump

It's a godsend for Emulator devs because they can now emulate GPUs using compute shaders instead of having to use the CPU

https://www.libretro.com/index.php/nintendo-64-vulkan-low-level-emulator-parallel-pre-alpha-release/

>>59149433
This sadly. AMD's TDP ratings are always way off mark. Can't wait until the benchmarks come out and we see this so-called 65W CPU get to over 100W.

>>59150086
IT WON'T BECAUSE IT WILL BE REACHING THE SAME TEMPS ON THAT ONE CORE YOU STILL HAVE ACTIVE.
Jesus christ how many times do I have to say so you fucking understand.
disabling cores for OC is done SOLELY FOR THE VRMs

>>59150083
You want to try shitposting elsewhere?

4k only comes in 60hz right now, and no single GPU can run AAA games much past 60fps anyways

>>59150050
multicore scaling in general isn't by much, i don't forsee it changing much in the future, as long as dx12 is os specific to a single version.

>>59149487
Yup. Intel is both cheaper and faster. R7 Ryzen is looking like a disaster right now.

>>59150096

Well yeah, its why is khronos needs to really push hard to catch up its documentation. Once the need for DX support vanishes so does MS's stranglehold on vidya - its why MSi s throwing its weight around trying to lock down windows even more precisely because it can see what could happen if they let their grip slip.

File: 1488200414225.png (68KB, 688x861px) Image search: [Google]

68KB, 688x861px

>>59150112
>AMD's TDP ratings are always way off mark
Not anymore.

>>59150116
>4k
Nobody plays games at 4k because 4k panels are shit and GPUs can't take it.
Plenty of people play games at high frame rate 1080p and 1440p, WHICH IS WHY THEY NEED A BETTER CPU.

>>59150094
Because if you do anything related the productivity the Ryzen chip will stop the 7700K, even something like recording your gameplay occasionally

>>59150146
>TDP is power usage :^)

>>59149853
Yeah not only high-end gaming, any kind of gaming or average desktop usage, Ryzen is going to fail.

AMD is doing the same shit they did before, add more cores and shitty real world performance.

>>59150115
>IT WON'T BECAUSE IT WILL BE REACHING THE SAME TEMPS ON THAT ONE CORE YOU STILL HAVE ACTIVE.
That's where you are wrong. Are you implying you will have the same overclocks on a stock cooler and a beefy cooler?

>>59150152
just fucking admit R7 is shit for gaming already

>>59150146
Now compare it to the 7700k. No one actually buys Intel's HEDT chips unless they're massive retards. Why bother benching against something no one will buy or is interested in buying?

>>59150148
Except the 7700K is only going to be about 15% faster at best and only if you have a $100 240mm AIO to throw on top of it, and only if you hit 5ghz stable with a more expensive Z270 board

where as with AMD you have the 1700 at 65W, and can even OC on a mid range B350 board

you may also need to delid the CPU, as intel as jews are cheaped out on the TiM.

some people have reported a 20C drop by replacing the TiM.

>>59150193
Nobody buys them because of the price

>>59150193
If things scale accordingly

r5 and r3 will still use less power than intel equivalents

>>59150146
Uh thanks for proving my point.

1700X is rated at 95W and it's hitting 123W

>>59150181
Jesus christ you're a fucking moron.

If you try to run a single 1700X core at 5GHz, that one core will generate just as much heat. Disabling cores won't increase the OC capabilities of the CPU, it will only let you do it without blowing up your motherboard VRMs

>>59150206
No one buys them because no one has a realistic need for more than 4c/8t. This is why Ryzen is such a shitty chip, you're paying for for performance you'll never actually use.

>>59150198
No, the 7700k is 15% faster than 1700 single core at stock speeds. And it currently costs less

>>59150245
>No one buys them because no one has a realistic need for more than 4c/8t
Most devs don't want to waste money on anything more right now

> This is why Ryzen is such a shitty chip, you're paying for for performance you'll never actually use.

>Ryzen is only 8c/16t chips :^)

>>59150245
This pretty much.

Really fucking sad how AMD is doing the same shit over and over again.

They win the benchmarks in video encoding and winrar tests.

Then get completely assblasted in gaming tests and any kind of regular desktop usage.

>>59150227
>Platform Power - Gaming ( CPU + GPU )
>Ryzen's power consumption is 25.57% lower

>>59150268
SAME IPC AS INTEL RETARD

>>59150260

>costs less

At select retailers making space in their inventory for the alpha money making Ryzen.

>>59150268
Except Ryzen is winning there too.

>>59150274
>1700X
>95w
>actually uses 130w

>>59150274
Go back and re-read the post you are replying to and think about how stupid you are.

>>59150227
its total power you mongoloid

>>59150234
Look, it seems you have trouble understanding what I'm conveying. This isn't about the max theoretical freqs a core can reach. It's about max freqs you can SUSTAIN without throttling. You can hit 4.5 Ghz on 8 cores but with a normal cooler you throttle after 2 mins, but with 4 cores you can sustain it in the same cooler.

>>59150260
It only costs less at Microcenter, don't know of anyone else like that, and it's been that price there for some time

>>59150281
>ignoring the rest of the post
cute

>>59150291
>>59150227
>motherboard, fans, etc. use no power
Thanks for clearing that up.

>>59150299
No it's not, look at the chart you complete moron.

File: 1488059530739.jpg (217KB, 800x1000px) Image search: [Google]

217KB, 800x1000px

>>59150263

>>59149507
TDP isn't power consumption

>>59150314
Yes it is, look at the chart you complete retard

File: AMD-Ryzen-7-1700-vs-Core-i7-7700K-DinoPC_GTA-V.png (2MB, 2560x1312px) Image search: [Google]

2MB, 2560x1312px

>>59150278
>>59150289

>>59150315
What are you going to say when AMD releases low core parts in the upcoming months? What argument will you use then?

>>59150334
poor little ryzen
killed before it even had a chance

>>59150334
In that graph, the overclocked 7700K loses to the stock 7700K

>>59150334
Thanks for proving their point. Intel only manages a few fps gain with much higher clocks.

>>59150299
The ryzen CPU is literally consuming 40W less when gaming than the intel one.

>>59150312
>>59150329
123W is CPU only you fucking idiots, for fucks sake learn to read.

AMD is going 28 watts over their spec just like they always do, lying about the power usage.

While the Intel CPU is hilariously going under their power spec.

>>59150339
they'll be shit compared to what intel is offering

>>59150334
>muh overclock

What is intel going to do about laptops were clocks can't go much past 3Ghz?

>>59150334
>

Where is this from? legit?

>>59150334
Fuck me anon, 3 fps lower than the intel chip at 5ghz, you sure got me pegged, it's a good thing you can't overclock the 1700

>>59150349
>ignoring min and max fps
AYMDRONE

>>59150312
VRM losses. Ridiculously efficient VRM would be about 90%.
A CPU actually using 90w would pull 100w across the VRM.

>>59150354
This is the important take away. The deltas in idle and load power between the two systems.

>>59150365
You need to try harder.

>>59150365
>While the Intel CPU is hilariously going under their power spec.
>going 50w over is somehow going hilariously under

>>59150370
IPC IS SAME YOU RETARD

>>59150379
Looking at the minimum fps makes Ryzen look even worse

>>59150365
>123W is CPU only you fucking idiots, for fucks sake learn to read.
Learn to read yourself, you fucking retard. It does NOT say CPU power. It says platform power.

>>59150387
>overclocking an intel part makes stuttering worse

LMFAO

Just bought a 7700K for $290 and a Z270 ASUS Gaming motherboard for $120
Ryzen can go suck my big brown cock

>>59150291
>6800K
>uses 127W vs 130W of 1700x
>with less cores and thus less performance
>somehow better

>>59150396
Holy shit you are illiterate

6800K is rated at 140W and it only hits 126W

damn, all the shills seem to be on overdrive now

>>59150414
u got rekt expect that to go down to 250 in a week.

The real question is will I find someone to buy my 6850k?

>>59150400
minimum fps =/= frame times

Just because you hit 20fps for a fraction of a second on one part of the benchmark doesn't mean the whole thing is unplayable

intel will price amd out of the market with current stock, then release a beast next year to finally put the war to bed. they needed a kick up the ass, since they literally haven't even been trying. now they actually have a reason to, we'll see some actual innovation. good times ahead.

>>59150405
Go re-read where it says "CPU ONLY"

File: やる夫.jpg (5KB, 192x150px)

5KB, 192x150px

>>59150129
>>Intel [...] cheaper
>6900K $1100
>1800X $500

>>59150435
Try craigslist and maybe facebook niggr

>>59150420
Oh, I thought you meant the gaming benchmark, since that's the one you talked about. It's the one Ryzen goes over 95w.

>>59150429
Nope. i5-7600K are going for $190 now, that's as low as they're ever going to go

File: ryzen-doa.png (177KB, 1089x677px) Image search: [Google]

177KB, 1089x677px

>>59150446

>>59149511
Joule say it does.
100% of the energy consumed by the thing turn into heat.

>>59150435
You have until thursday before your chip becomes worthless

>>59149042
jew spotted

>>59150420
ask yourself - why does 6c/12t intel consumes same power as 8c/16t amd?

>>59150441
That means they stressing the CPU and the GPU is idle, idiot.

File: 1462252897511.png (211KB, 327x316px) Image search: [Google]

211KB, 327x316px

>>59150467
>that's as low as they're ever going to go

The r3 1200 will be 129 usd

>>59150477
>6900k=7700k
???

>>59150441
Can you read m8? Is it difficult for you? Or don't you just comprehend?

>>59150448
Yea probably going with that, ebays bullshit now.

>>59150365
technically they don't "lie" because they explain they measure TDP differently but it is indeed bullshit

I expect Ryzen 65W CPUs to reach 100W easily

>>59149489
There is no performance or lifespan difference between running your CPU at 25C with phase-change cooling vs 80C with the stock cooler.
Lower temperatures are a meme for bragging on reddit.

Your motherboard, RAM, harddrive/SSD, and powersupply will fail multiple times before your CPU does.

>>59150494
And Kaby Lake can hit higher clocks, have higher IPC, a more robust memory controller, etc.

Your point being?

>>59150506
lower temps mean more overclocking headroom

>>59150477
>$310 7700k
Yeah, there's literally no reason to buy a ryzen now. Thanks for the price drop though, ayymd.

>>59150516
>And Kaby Lake can hit higher clocks,

Still unconfired but doesn't matter in laptops where the majority of consumer sales lie LOL

>have higher IPC

NOPE LOL

>>59150524
Nigga, it's $290 in some retailers.

>>59150524
Pretty much.

I can't see any reasonable reason to buy it except for extreme multi-threaded apps which is a really niche case.

Are people in this thread actually saying they'd rather have 7700k than 6900k?

>>59150539
>more robust memory controller
>No ECC

KEK

File: bang for buck.png (193KB, 877x496px) Image search: [Google]

193KB, 877x496px

>>59150477
wait for 1800x and 6950x..

>>59150548
no, they're saying they'd rather have a 7700K than an R7 1700

which makes sense

>>59150548
There is literally zero reason for anyone to go with a half-assed piece of shit CPU like Ryzen.

>>59150557
AM4 does not support ECC RAM, so...

>>59150544
This is what happen when you don't have a monopoly anymore.

>>59150548
>6900k
Why should I need it when I mostly stick to playing vidya?

File: hurrrr.png (26KB, 628x677px) Image search: [Google]

26KB, 628x677px

>4c/8t is more than enough

Is the only thing you faggots care about gayming?

>>59149690
>>59149604
jew got BTFO

>jewtel damage control thread

File: 6900k-7700k.png (242KB, 1093x780px) Image search: [Google]

242KB, 1093x780px

>>59150548
6900K gets beat by the 7700K in gaming and desktop use

Just like Ryzen, 6900K is a niche cpu only good for highly multi-threaded apps

I'd take 7700K any day of the week

>>59150547
>>59150524

You're going to regret your purchase once the r5s and r3s come out

>>59150567
Stop having buyers remorse.

>>59150575
You don't run any programs along with the games? Also you don't see games using more than 4 cores in the future?
This is literally a question of 7700k vs 6900k for THE SAME PRICE. 1700 will also be much cheaper to overclock to go PAST 6900k.

>>59150614
It will be interesting to see what the R5 and R3 looks like

So far the R7 looks like a failure though

So basically for vidya, wait for r5 and laugh.

>>59150628
>failure
that's because you retards keep on comparing it to the wrong processor
>HURR MUH 8 CORE AIN'T CLOCKING LIKE BABBY LAKE

>>59150616
Buyer's remorse? I've got a 7700K, the hands-down BEST overclocking silicon on the planet, for less than a shitty, overhyped and underperforming Ryzern 1700X.
Better performance
Better overclocks
Better cost
And just better at everything
Enjoy your shitty dual module "eight core" marketing gimmick

>>59150614
probably not since GPUs are the bottlenecks in most scenarios anyway.

>>59150628
>beating intel's comparable chips for half the price
>failure

also

>bringing down prices and making the market fair again
>failure

???

>>59150198
you don't need an aio or a tim swap, just a nice air cooler. running a cpu at anything around 80 or below is fine

>>59150651
There's no reason to believe that the AMD 4c/8t offerings aren't going to be half the price of your babby lake

>>59150648
>The only reason someone would want AMD to fail is if they work for Intel, or if they just bought an i7 and are having buyers remorse, but even then they should be happy that there will be a leap in progress and future upgrades will be cheaper.

But please remember to make threads 2 years from now saying how you shouldn't have fallen for Intel now that games are starting to benefit from 8 cores. I will be laughing.

>>59150628
lga2011v3 is a failure too

>>59150614
>r5s and r3s
>wait for cpus that are slightly better than current low-end i5s
Yeah I'd rather not have a cpu that will bottleneck everything in 3 years.

>>59150502
No, 65w Ryzen pulls 65w under full load.
95w Ryzen pulls 95w under full load.

There is a huge difference between measuring what is actually being pulled at the socket and consumed by the CPU, and what a power supply is pulling, or what is being pulled at the rail.
Power supplies have their own efficiency losses, and voltage regulators have efficiency losses.

If you actually isolate CPU power draw then you'll see the i7 6800k will pull about 140w under a heavy load on all cores. Every chip in the Broadwell-E line has clocks set to utilize their full TDP under heavy load. Under less intensive loads, even if you're using all cores, power draw will be 100-110w. Just because a workload is utilizing all cores doesn't mean its as intense as Prime95.

The chink review shows the Ryzen system pulling 123w under the CPU only metric, this is very clearly taken from the wall, or taken from the rail. It shows 126.87w for the 6800k.
The gaming CPU+GPU power figures are 154.66w for Ryzen, and 194.2w for the Broadwell-E system. Same GPU, now a significantly larger delta between the two.
Even in the office/productivity workloads we see over a 30w delta between the two.
There is a very clear distinction in power between the two systems, and at best these power figures were taken from a rail without factoring any losses.

Its astounding how little a tech board understands about tech.

>>59150701
>wait for cpus that are slightly better than current low-end i5s
so like 4 core i7's?

>>59150334
high spikes just inflate average fps, aka amd have better gameplay performance.

>>59150681
Intel will drop prices to make them competitive anyway. If the 7s are barely competing with kaby lake at a higher price, what makes you think the 5s will>?

>>59150638
>>59150653
>ryzen not even out yet
>AMDrones already claiming it's a success based off on AMD sponsored benchmarks
March 2nd and actual real benchmarks can't come soon enough.

File: VomitChan.jpg (56KB, 386x460px) Image search: [Google]

56KB, 386x460px

>>59150657
>80 or below is fine

>>59150737
>intel
>dropping prices on relatively new products
even from a PR standpoint, it's a retarded decision
people will realize that they're being priced gouged

File: INTEL BTFO.png (245KB, 800x612px) Image search: [Google]

245KB, 800x612px

>>59149042

>>59150757
March 2 is when the chips go on sale.
Review embargo is lifted tomorrow on the 28th.

>>59150724
>7600k and 7700k
>not cpus with extremely high future-proofing

>>59150757
yes, and I'll get to see you intel babbies cry

>>59150769
>people will realize that they're being priced gouged
They should have realized this a long time ago, but AMD dropped the ball with bulldozer

>>59150737
>Intel will drop prices
Isn't that good then? This isn't about AMD vs Intel you fucking fanboys, it's about BREAKING the inflated monopoly prices Intel's been charging, and making the competition real again.

>>59150757
>AMDrones
You really are a little child having an irrational defence for a company that has never done anything for you, aren't you?
How hard is it to see that AMD's success is good for consumers?

>>59150737
>If the 7s are barely competing with kaby lake at a higher price
that's because you still don't understand that 1700/1800x is competing with 6900/6950x and not with 7700k

>>59150809
so they're competing with a niche market that barely exists?

>>59150819
That won't be the case soon, since AMD just made 8 core chips affordable.
More people using 6-8 core chips -> more games and software taking advantage of them.

>>59150713

If these results hold true watch as /g/ will (once again) say power draw doesn't matter. When fermi was a nuclear inferno power draw didn't matter until it suddenly did when Nvidia got a lead over ati/AMD with it. That what is happening here, only in reverse.

>>59150819
I dunno about you but every machine I own that isn't a laptop has 8+ cores.

But I also don't play video games in my mom's basement all day.

>>59150819
it's niche because of intels prices, now that's it available and more and more engines can utilise more and more threads it will be mainstream soon, don't worry

>>59150819
chicken/egg
is broadwell-e niche because it's cost-prohibitive or is it cost-prohibitive intel has to make adequate returns on a niche market?
AMD seems to be confident going with the former.

>>59150713
>65w Ryzen pulls 65w under full load.
>95w Ryzen pulls 95w under full load.

But it won't, that's the whole argument here. I wasn't referring to the chink review, in fact I hadn't even seen that post. I was simply stating that it is a fact that AMD measures TDP differently from Intel (they themselves explain it on their website), and an Intel CPU rated at 65W TDP will at most reach ~70W during a gaming load, whereas a typical FX CPU rated at 95W will reach over 125W easily. That's why I doubt Ryzen will be any different.

>>59148945
>4/8 is more than enough
Nobody says "a Toyota is more than enough" if they can get a porsche for the same price. I don't believe you. Go away shill.

>>59150898
>for the same price
but OP clearly said the Toyota is cheaper, you idiot

>>59150916
you still have to buy a cooler for the i7

>>59148945
While you are right, Intel will not make a leap ahead in a single-threaded performance in the near future: it requires a new architecture, something more advanced than "Core i" and it was in the game for 10 years now. There were no plans announced for such development. HOWEVER, Intel can add more cores to compete with Ryzen, which is 2x more capable in multi-threading.
What does that mean for us?
- More cores. 8- and 10- core consumer CPUs will be there in a few years.
- Intel CPUs will support AVX2 while AMD CPUs will be cheaper.
- AMD CPUs are more futureproof as of now.
The only thing Intel can bring to the table is the Power8 ST licensed technology, when two cores work together to improve a ST performance.

>>59150882
Don't try to bullshit someone far more informed than you, kid.
Every CPU on the market is allowed to violate TDP for brief periods, particularly mobile chips, but what you're saying is nothing but a bold faced lie.
http://www.tomshardware.com/reviews/intel-kaby-lake-core-i7-7700k-i7-7700-i5-7600k-i5-7600,4870-8.html

Under a nominal heavy workload the Kaby Lake i7 7700K will pull 21.5w per core at 4.2ghz. It'll routinely pull just under 100w from a strictly CPU load.
However if you really hammer them in intel's own testing utility they'll pull 30w per core. The Kaby Lake i7 7700k will pull 135w, without even touching the IGP. Strictly core logic and uncore. From a chip with a 91w TDP.
These figures are totally isolated CPU power consumption. That is what the CPU itself is consuming, not the board, not from the wall, that is pure power draw of the CPU under load.

95w Ryzen pulls 95w.

>>59148945
Then don't. Buy the Intel. No one fucking cares.

>>59150975
>The only thing Intel can bring to the table is the Power8 ST licensed technology, when two cores work together to improve a ST performance.

You're not even thinking of the right company, and you clearly know absolutely nothing about IBM's POWER8.

>>59151000
dude your attempts to sound superior are hilarious, but you clearly have no idea what you're talking about

keep believing Ryzen will pull 95w if you want. Just get literally any previous AMD processor and see if it matches its rated TDP. Or, you know, you could simply read my previous 2 posts and finally understand that AMD l-i-t-e-r-a-l-l-y states that TDP is measured differently from Intel, and that it is expected behaviour to reach values well beyond its rated TDP, unlike on Intel chips

>>59151069
Or you could just accept that no-one is accepting your bullshit, intelshill.

>>59151069
>talk out of your ass, lie, and shitpost
>get BTFO by sourced facts
>keep shitposting like the pathetic little kid you are

http://www.tomshardware.com/reviews/amd-fx-8370e-cpu,3929-2.html

95w rated FX8370e pulling 75w at the rail on average, 65w~ actual CPU draw when VRM losses are accounted for.

>>59151069
AMD cpus have configurable tdp in the uefi...

>>59151000
>Every CPU on the market is allowed to violate TDP for brief periods

this is the most ridiculous sentence I've ever read

>>59151094
>>59151102
>>59151112
http://www.intel.com/content/dam/doc/white-paper/resources-xeon-measuring-processor-power-paper.pdf

>>59150916
He said it's cheaper, but it's not.

>>59151206
it is in europe

File: 04-Power-Consumption-PTU.png (24KB, 711x534px) Image search: [Google]

24KB, 711x534px

>>59151156
Thats a simple fact. Why even bother posting when you're so entirely clueless?
TDP is not an ultimate limit. Chips can exceed TDP for set periods. Mobile chips tend to have time frames they can exceed TDP for before they throttle.
Educate yourself instead of talking out of your ass.

>>5915117
Try again, little kid.
http://www.tomshardware.com/reviews/intel-kaby-lake-core-i7-7700k-i7-7700-i5-7600k-i5-7600,4870-8.html

Second time I've linked this. If you can't refute the facts then don't post.
91w Kaby Lake pulling 135w with just the CPU alone, IGP is not being utilized.

>>59151174
>linking intel's website
>not anything like, say, actual measures of AMD CPU's powerdraw that would show how wrong he is
Kill yourself intelshill.

>>59151225
I'm imagining an obese american 15 year old posting this

>>59151102
calling people kid does nothing for your argument.

>>59151213
>implying Europastan counts for anything anymore.

>>59151250
>I have no argument
>I will now act like I'm offended to seem like an enlightened individual without making a point

>>59151250
Act like a kid, and you deserve to get treated like one.
I'm posting facts and providing sources while a little kid just talks out of his ass and lies.

>>59151271
I'm not the one you're arguing with. In fact I agree with your points. I'm just saying calling people kid makes you come off like a basement dweller.

Ahhh... Marvin@mykancolle.

What a great source!

>>59151225
what are you even talking about though? We're talking heavy load, not maximum load. You find me an Intel CPU that reaches more than 5W of its rated TDP under a heavy load achievable without the Power Thermal Utility. And then look for an AMD processor rated at the same TDP under a similar load.

I don't even know what you're trying to achieve here. It's a fact that Intel measures TDP in a way that it will give you plenty of head room, whereas AMD measures it as a sort of average usage under a normal load. I'm not making this up, one of the biggest (if not THE biggest) problem with AMD processors has ALWAYS been power consumption. If you're not trolling then you either don't know what you're talking about or are actually mentally deranged

>>59151361
Its beyond pathetic that you're still talking out of your ass. You're proving beyond all doubt that you're just dedicated to lying and shitposting instead of educating yourself.
The very same link in that post shows the 91w Kaby Lake i7 7700k pulling just shy of 100w under a normal FPU workload. Again, IGP not being utilized. That is strictly core and uncore drawing that much power. It is violating TDP under a normal CPU workload. I chose this one desktop part as an example because the mobile parts are even worse when the CPU+IGP are utilized at the same time. Its the same for CoreM all the way up the line for their mobile SKUs. The Iris Pro Haswell chips in the old MBP could pull almost 90w, well above their already excessive 47w rated TDP.
5w rated Broadwell CoreM chips were pulling up to 17w under load.
Both examples amply shown in numerous notebookcheck reviews.

Your statement that intel chips stay within their TDP at all times is simply an absolute fabrication.

Protip: In the future if you don't want to get called out for lying, just don't lie in the first place.

>>59151519
http://valid.x86.fr/x4e04a
This is an intel chip not staying within TDP. In fact it's running faster than all but 13 Core i7-4810MQ's Hell, two of them on that list arn't even real 4810MQ's they're 4900MQ's.

So... You're right.

>>59151519
holy shit what are you even talking about

this is incredible

>>59148945
>passmark

File: SHITS.ON.FIRE.YO.png (45KB, 1269x678px) Image search: [Google]

45KB, 1269x678px

Anyone want to put out this fire?

File: NIGGA.GET.THE.WATER.png (48KB, 1264x692px) Image search: [Google]

48KB, 1264x692px

GET THE WATER.

>>59150567
Yes it does support ECC. You just need a motherboard that does aswell.

>>59151615
>get utterly blown the fuck out
>do nothing but shitpost

>>59148945
POOZEN DOA

>>59149042
You embarrassed yourself, or false flag

>>59149900
>Unless you have a meme 144hz monitor the 1700 is overkill for any game,

There's been an influx of CPU bounds games lately, GTA V being one of the most notorious examples.

I love this meme with "x cores is enough"
Who needs dual-core when one is powerfull enough?
Who needs more than a dual-core? Quad-cores are too much.
Now we are at "quad-core+HP is more than enough".

I really hope Ryzen will be good, my poor i5 Ivy is starting to fall behind.

>>59151246
nice ad hominem retard

>>59149294
I'm almost certain that some reviewers will do tests of disabling 2 or 4 cores to emulate a 1600X and 1400X on an 1800X. (hopefully they do the 1400X one correctly by disabling an entire CCX so there's not 2 cores using too much cache).

The 1600X should actually have the best IPC of them all, more than the 1800X because it has 33% more cache per core. 8mb shared between 3 cores is a lot.

>>59152864
They all have the exact same IPC.
6 core Ryzen chips do not have the same amount of cache as the 8 core chips.
The L3 is a victim cache.

>>59149795
Quit fooling yourself.

4c/8t might be enough for 2017, but you have to be a moron to buy something that's not still going to be good a few years from now. You save money future proofing yourself instead of repeatedly doing upgrades every 2 years.

At least you're not quite as stupid as the people that think 4c/4t is enough, I'll give you that.

But look in 2013, people still thought 4c/4t was enough. Now days, it clearly isn't. Why the fuck do you think 4c/8t will still be enough in 2020? It makes no god damned sense. History repeats itself whether you idiots learn or not.

File: 1440542150252.jpg (23KB, 320x418px) Image search: [Google]

23KB, 320x418px

>AM4 support till 2020

This alone should be enough reason to buy a Ryzen to be honest. Intel likes to fuck around with sockets too much, it's revolting.

>>59152920

>6 core Ryzen chips do not have the same amount of cache as the 8 core chips.

Given how previoussly AMD cut down 8 to 6 I would not agree with this.

File: 1487463821177.png (2MB, 1440x900px) Image search: [Google]

2MB, 1440x900px

>>59150365
>>59150502
>>59150329
>>59150291
>>59150274
>>59150227
>>59150146
That image is transcribed wrong.
The 8 core Ryzen more hilariously beats the 6 core 6800k in performance/watt even more than that.

1700X vs 6800k
>平台功耗-CPU满载(越小越好) : Platform power consumption - CPU full load (the smaller the better)
123 vs 126.87
>平台功耗-待机(越小越好) : Platform power consumption - standby (the smaller the better)
62.77 vs 98.74
>平台功耗-游戏(越小越好) : Platform power consumption - Game (the smaller the better)
154.66 vs 194.2
>平台功耗-办公(越小越好) : Platform power consumption - office (the smaller the better)
81.55 vs 113.5

So 25-40% more power efficient, while it's 15-40% more powerful in workstation tasks and still about 5-10% on faster average faster in gaming.

And as far as >going over their spec
Yes, as many said, it's platform total power measured from the wall. You can't really trust software power monitors to correctly monitor a specific part when it's one connection going to the motherboard.
Not only does the translation say "platform power consumption" but you can see how much gaming jumps up the power usage.
Also, TDP is heat dissipation needed, and not necessarily how much wattage a chip draws.

There you go. Bunch of a fucking morons making me waste time educating you. Why can't you educate yourselves?

>>59150576
Yeah, but many of them are in denial about it.

>I-I once rendered an animation of Midna and my 4 cores and 4 threads were good enough okay

>>59153155
The L3 is a victim cache.
Data is only written to the L3 if it is first written to the L2.
L2 is private per core.
The L3 is not monolithic blocks, it is divided into 4 slices.
Each slice defacto belongs to a single core, and other cores only access the other slices through a buffer.
This buffer provides all cores with the *same average latency* when accessing any other slice.

>>59149743

Enjoy your Collosal mistake and Gut-Wrenching Jealousness

File: 20170227095022.png (22KB, 1095x805px) Image search: [Google]

22KB, 1095x805px

simple:

>>59150611
>even that says the 6900k is better in gaming
lmao

>>59150616
Don't bother. None of these children understand anything about computers and how multithreading and schedulers work.

They don't understand how benchmarks are done on sterile, fresh systems with nothing running in the background.

And they'll never understand, because most of them are stupid teenagers. Just have to give them 10 years to grow up and they might be able to read and understand technical information by then.
And others of them are just people with buyer's remorse.

>>59153290
>HTTP:\\
>\\
What the fuck

>>59151696
>>59151709

intel cucks will ignore this benchmark

>>59153290
>multithreaded singlethread performance x64
I admit I laughed.

>>59151246
So an obese American 15 year old is smarter than you?

I feel sorry for your self esteem.

>>59150787
>mfw no one ever in the next 1000 years will ever need more then 4 cores or 8 threads outside of really stupid niche shit.

Thats it boiz, AMD dead keep buying 4 core i7.

>>59148945
Ryzen is better

>>59150315
>RELEASE SKYLAKE AGAIN

>>59150764
80C stable is fine for Intel CPUs. Lower temps just give you more headroom for overclocking or peace of mind if you live in Phoenix and you're afraid your AC will malfunction.

>>59150334
>Averages all exactly the same
>everyone keeps looking at the green bars.

Fucking, everyone need to KILL THEMSELVES.

>>59152989
>AM4 support till 2020
B U L L S H I T
get ready for your AM4+ next year

>>59154948
AM3 and AM3+ are compatible.

This then means nothing. Enjoy Intel releasing rebrand CPUs that require totally new sockets.

>>59154948
t. mind fucked with stockholm cuck syndrome thanks to Intelaviv and can't see any other way

>>59149309
they do it for planned obsolescence

never buy from kikes

>>59151225
It's hilarious seeing a Core M being rated at 4.5W SDP. It will happily use 21W for a few seconds before throttling down to about 10-12W. TDP and SDP counts are really poor ways of gauging power consumption.

>>59149581
>>59149532

Note: Intel, or at least Israel proper, IS paying people to post things like this. Make a small youtube channel critical of yids in some way and watch the bullshit comments come pouring in. The jews and other liars need to be forcefully excluded from western society.

Or we'll keep getting shit like this.*

>>59149487
all RYZEN cpus destroy the 7700k ... single thread is fucking history . multi thread is present and future .

>>59153944
Holy fuck you are retarded. The averages are not the same.

File: 772.jpg (42KB, 800x587px) Image search: [Google]

42KB, 800x587px

>I5 7500 and 7600 don't come with stock cooler

Well fuck you too Intel, I'll get a Ryzen instead

>>59155659
lmao?

Why would a non-k not come with a cooler. That makes no sense.

Holy fuck that makes Ryzen even better of a value. Fucking Jews. It's not just the shitty TIM.

>>59155679
>Why would a non-k not come with a cooler. That makes no sense.

My i5 4570 came with a cooler, though.

>Holy fuck that makes Ryzen even better of a value. Fucking Jews. It's not just the shitty TIM.

https://www.youtube.com/watch?v=9AVZ_x64hg4

and it looks damn fine too

>>59155792
>My i5 4570 came with a cooler, though.
I don't understand how this is relevant to a 7500 and 7600 not having one.